SQL joining tables including NULL counts - sql

Question

This I'm hoping is an easy fix, I have 2 tables, one of days over a 6 month period, the other with sitenames, day(date) and count of attendances that day.
I'm wanting to create a table where for each site, it has a row for every day in the 6 month period and takes that sites count which corresponds to the day, but I also want it to show where there is a NULL (no attendance on that day) - I can do it where it brings out only days with attendance but not the other way around. :(
Example data here: NOTE, the data is held in two temporary tables
Date table #Data
CallDate rn
2022-08-01 1
2022-08-02 2
2022-08-03 3
2022-08-04 4
2022-08-05 5
2022-08-06 6
2022-08-07 7
2022-08-08 8
Attendance table: #SiteData
SiteName CallDate Count
Bassetlaw 2022-08-30 1
Bassetlaw 2022-08-31 1
Bassetlaw 2022-09-13 3
Bassetlaw 2022-09-15 5
Bassetlaw 2022-09-23 1
Bassetlaw 2022-09-27 1
Bassetlaw 2022-11-21 1
Bassetlaw 2022-11-23 1
Bassetlaw 2022-11-26 1
Bassetlaw 2022-11-28 1
So in this instance, I would have 6 months worth of rows, but only 10 days worth of data. I need NULLs for the other days, not just 8 rows.
NOTE: There are more sites, I would want this repeated for all site. In essence, I want a table that has all sites with a row per site per day for 6 months irrespective if they had an attendance or not.

Answer 1

This is done by using the LEFT JOIN command.
See here: http://sqlfiddle.com/#!18/0218c/2
CREATE TABLE T_Data (
rn int,
CallDate date
);
CREATE TABLE T_SiteData (
CallDate date,
Amount int
);
INSERT INTO T_Data SELECT 1, '2022-08-01';
INSERT INTO T_Data SELECT 2, '2022-08-02';
INSERT INTO T_Data SELECT 3, '2022-08-03';
INSERT INTO T_Data SELECT 4, '2022-08-04';
INSERT INTO T_Data SELECT 5, '2022-08-05';
INSERT INTO T_Data SELECT 6, '2022-08-06';
INSERT INTO T_Data SELECT 7, '2022-08-07';
INSERT INTO T_Data SELECT 8, '2022-08-08';
INSERT INTO T_Data SELECT 10, '2022-08-09';
INSERT INTO T_Data SELECT 11, '2022-08-10';
INSERT INTO T_Data SELECT 12, '2022-08-11';
INSERT INTO T_Data SELECT 13, '2022-08-12';
INSERT INTO T_Data SELECT 14, '2022-08-13';
INSERT INTO T_Data SELECT 15, '2022-08-14';
INSERT INTO T_SiteData SELECT '2022-08-01', 1;
INSERT INTO T_SiteData SELECT '2022-08-03', 1;
INSERT INTO T_SiteData SELECT '2022-08-05', 3;
INSERT INTO T_SiteData SELECT '2022-08-12', 5;
SELECT
d.*,
sd.Amount AS [Count]
FROM
T_Data AS d
LEFT JOIN
T_SiteData AS sd
ON
sd.CallDate = d.CallDate
This returns the following:
rn CallDate Count
1 2022-08-01 1
2 2022-08-02 (null)
3 2022-08-03 1
4 2022-08-04 (null)
5 2022-08-05 3
6 2022-08-06 (null)
7 2022-08-07 (null)
8 2022-08-08 (null)
10 2022-08-09 (null)
11 2022-08-10 (null)
12 2022-08-11 (null)
13 2022-08-12 5
14 2022-08-13 (null)
15 2022-08-14 (null)
I used the MS SQL server syntax. But the SELECT query should be the same on almost any other DBMS.
If the table SiteData contains more than one entry per CallDate, the query needs to be adjusted. But your example only showed one entry per CallDate.